Struct Model 

pub struct Model<'a> {
    pub arg_type_id: TypeId,
    pub ret_type_id: TypeId,
    pub data: NormalInverseWishart,
    pub ctxt: &'a Context,
}

A representation of the information known via Bayesian regression of a nonlinear model from the compressed space of the given argument TypeId to the compressed space of the given return TypeId. The wrapped NormalInverseWishart linear model is defined on mappings from the feature space of the input to the compressed space of the output, and the whole Model is assumed to exist in the given Context.

Fields

arg_type_id: TypeId

ret_type_id: TypeId

data: NormalInverseWishart

ctxt: &'a Context

Implementations

impl<'a> Model<'a>

pub fn get_total_dims(&self) -> usize

Gets the total number of coefficients used to define the mean of this Model

impl<'a> Model<'a>

pub fn get_context(&self) -> &'a Context

Gets the Context that this Model exists within.

pub fn sample(&self, rng: &mut ThreadRng) -> Array2<f32>

Draws a sample of a linear mapping from the feature space of the input to the compressed space of the output from the distribution defined by this Model.

pub fn sample_as_vec(&self, rng: &mut ThreadRng) -> Array1<f32>

Identical to Self::sample, but the result is flattened.

pub fn get_mean_as_vec(&self) -> ArrayView1<'_, f32>

Gets the mean of the underlying NormalInverseWishart model from the feature space of the input space to the compressed space of the output, as a flattened vector.

pub fn get_inverse_schmear(&self) -> FuncInverseSchmear

Gets the FuncInverseSchmear for the underlying NormalInverseWishart model from the feature space of the input to the compressed space of the output.

pub fn get_schmear(&self) -> FuncSchmear

Gets the FuncSchmear for the underlying NormalInverseWishart model from the feature space of the input to the compressed space of the output.

impl<'a> Model<'a>

pub fn new( prior_spec: &dyn PriorSpecification, arg_type_id: TypeId, ret_type_id: TypeId, ctxt: &'a Context, ) -> Model<'a>

Constructs a new Model with the given PriorSpecification for the underlying NormalInverseWishart model, the given argument and return types, and existing within the given Context.

Trait Implementations

impl<'a> AddAssign<&NormalInverseWishart> for Model<'a>

fn add_assign(&mut self, other: &NormalInverseWishart)

Updates this Model to reflect new prior information from the given NormalInverseWishart, which is assumed to come from another Model with the same input and return spaces.

impl<'a> AddAssign<DataPoint> for Model<'a>

fn add_assign(&mut self, other: DataPoint)

Updates this Model to reflect new regression information from the given DataPoint, which is assumed to map from the compressed space of the input to the compressed space of the output.

impl<'a> AddAssign<DataPoints> for Model<'a>

fn add_assign(&mut self, other: DataPoints)

Updates this Model to reflect new regression information from the given DataPoints, which is assumed to map from the compressed space of the input to the compressed space of the output.

impl<'a> Clone for Model<'a>

fn clone(&self) -> Model<'a>

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl<'a> SubAssign<&NormalInverseWishart> for Model<'a>

fn sub_assign(&mut self, other: &NormalInverseWishart)

Undoes the action of the corresponding add_assign method.

impl<'a> SubAssign<DataPoint> for Model<'a>

fn sub_assign(&mut self, other: DataPoint)

Undoes the action of the corresponding add_assign method.